The incidence of preterm births (infants born at less than 37 weeks of gestation) continues to increase, and identifying approaches that optimize weight gain and reduce length of stay in these babies remains one of the greatest challenges in current neonatal practice. Recent work from a variety of sources provides promising evidence that non-invasive programs of "assisted exercise" (gentle flexion, extension, and passive range of motion maneuvers) in preterm infants can hasten body mass increase and improve bone strength. However, prospective studies on assisted exercise have been limited in sample size and have not included a single set of anatomic and biochemical mediators of growth that could shed light on underlying mechanisms and pave the way for clinical application. A series of cross-sectional and prospective studies are planned to determine: 1) the fundamental relationships between body composition and growth/inflammatory mediators in a cross sectional cohort of healthy preterm (reaching term) and term infants; and 2) the effect of 4 weeks of assisted exercise training in otherwise healthy, growing, preterm infants using a prospective, randomized control study on: a) weight gain, b) muscle mass, c) bone strength, d) key circulating growth and stress/inflammatory mediators, and e) spontaneous physical activity post exercise. State-of-the-art, minimally invasive techniques using bone and muscle ultrasound will be used to assess body composition. The anatomic measurements will be complemented by measuring energy intake and by examining circulating mediators of two antagonistic systems that are known to influence growth early in life, namely, the growth hormone-insulin-like growth factor axis (GH-IGF-I) on the one hand, and stress/inflammatory cytokines, on the other. GH-IGF-I will be assessed with IGF-I (free and bound) and GH binding protein, giving insight into both overall GH activation and the degree of GH insensitivity. Stress/inflammatory activity will be assessed with interleukin-6 and interleukin-1 receptor antagonist giving insight into both inflammatory and anti-inflammatory elements of the stress response. Finally, the study has been designed to minimize problems previously identified in this type of research (e.g., performance bias, blinding of outcome measures, uniformity in the intervention, lack of proper control) by creating an independent study observation and outcome measurement team under the auspices of the institute on General Clinical Research Center. [unreadable] [unreadable]